Submitted to: Medicine and Science in Sports and Exercise
Publication Type: Peer reviewed journal
Publication Acceptance Date: 8/25/2006
Publication Date: 1/2/2007
Citation: Anderson, C.B., Masse, L.C., Hergenroeder, A.C. 2007. Factorial and construct validity of the athletic identity questionnaire for adolescents. Medicine and Science in Sports and Exercise. 39(1):59-69. Interpretive Summary: Maintaining exercise and physical activity over long periods of time and over one’s lifetime likely reflects one’s self-concept or identity as an active person. Measurement instruments are needed to measure such aspects of self-concept so that people at risk for inactivity can be identified and changes in self-concept, perhaps as a result of an intervention, can be documented. This research developed a questionnaire to measure athletic identity in adolescents. The questionnaire has 40 questions that tap four general themes: athletic appearance, competence in physical activities, importance of physical activities, and encouragement from parents, teachers, and friends. Few instruments exist that measure self-concept in regard to exercise, sport, and physical activity behavior, especially in adolescents and children. This work produced a questionnaire that adolescents can complete on their own in various settings, such as in a school classroom. It was developed using advanced statistical techniques, which assure that the measure is reliable and valid.
Technical Abstract: This research describes the development of a measure of the general attribute of "athletic" in adolescents, encompassing exercise, sport, and physical activity. Based on a theoretical model supported in adults, the 40-item Athletic Identity Questionnaire (AIQ) for adolescents assesses four dimensions: appearance, competence, importance of activity, and encouragement from three sources: parents, friends, and teachers/other adults. Structural equation modeling was used to evaluate the hypothesized four-factor model in a development sample of 408 adolescents in eighth grade (mean age 13.4 yr). A separate sample (N=1586) was used to cross-validate the final model. Construct validity was examined by testing the model's relationship to self-reported (Modifiable Activity Questionnaire-Adolescent, Previous Day Physical Activity Recall, Youth Risk Behavior Survey) and objectively measured physical activity (MTI accelerometer in sample 3, N = 100). Confirmatory factor analysis supported the four-factor structure, and there was also support for a higher-order model. LISREL correlations between the AIQ factors and self-reported physical activity ranged from 0.32 to 0.61, TV watching from -0.20 to -0.50, and sport-team participation from 0.20 to 0.54. Pearson correlations between the AIQ factors and MTI vigorous physical activity ranged from 0.09 to 0.26 and MTI moderate from -0.06 - 0.22. Findings support the factorial and construct validity of the AIQ for adolescents.